The mRNA expression profiles were obtained from Gene Expression Omnibus (GEO) dataset website (ncbi.nlm.nih.gov/geo/). GSE12452 was used to analyze the mRNA expression of FOXG1 in NPC tissues and normal tissues. FOXG1 mRNA expression data was processed using the R software (3.4.0 version, r-project.org/) and limma R package (20).

A total of 70 NPC tumor tissues and matching para-carcinoma tissues were obtained from the Department of Otolaryngology-Head and Neck Surgery, The Second Affiliated Hospital of Kunming Medical University between January 2018 and March 2020. The inclusion criteria were as follows: i) Patients had never received radiotherapy or chemotherapy before surgery; ii) patients had no medical history of other malignant tumors; and iii) the diagnosis of all samples was confirmed by histopathology of NPC. The exclusion criteria were: i) Patients had incomplete clinicopathological data; and ii) patients were unwilling to cooperate with treatment. The collected samples were immediately frozen in liquid nitrogen after surgical resection and stored at −80°C until use in subsequent assays. The study was approved by the Ethics Committee of The Second Affiliated Hospital of Kunming Medical University (approval no. KYDE201801012), and all patients (20 females and 50 males; age, 28–75 years) signed informed consent forms.

First, 10% neutral buffered formalin was used to fix the tissue samples for 24 h at room temperature, which were then dehydrated and embedded in paraffin wax. Next, the paraffin-embedded samples were cut into 4-µm thick sections, and the sections were dewaxed with xylene. They were then dehydrated with gradient ethanol, and endogenous enzymes were removed with 3% H₂O₂ for 10 min at room temperature. Subsequently, the sections were treated with 10 mM Tris-EDTA (Thermo Fisher Scientific, Inc.) at 125°C in a pressure cooker for antigenic retrieval and then incubated with primary antibody (FOXG1; 1:200; cat. no. ab196868; Abcam) overnight at 4°C, followed by incubation with goat anti-rabbit secondary antibody (1:250; cat. no. ab150081; Abcam) for 30 min at room temperature. Finally, the sections were stained with 3, 3′-diaminobenzidine (Thermo Fisher Scientific, Inc.) for 5 min at room temperature, counterstained with hematoxylin for 3 min at room temperature and mounted with neutral gum. Images were captured using an Olympus FV1000 laser scanning confocal microscope (Olympus Corporation; magnification, ×100 and ×400).

Human immortalized nasopharyngeal epithelial cells (NP69; BeNa Culture Collection; Beijing Beina Chunglian Institute of Biotechnology) were cultured in a keratinocyte/serum-free medium (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 0.2 ng/ml human recombinant epidermal growth factor, 2% FBS (Gibco; Thermo Fisher Scientific, Inc.) and 1% streptomycin and penicillin (Gibco; Thermo Fisher Scientific, Inc.). The NPC cells (C666-1 and SUNE-1; The Cell Bank of Type Culture Collection of Chinese Academy of Sciences) were cultured in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.) containing 5% FBS and 1% streptomycin and penicillin. Both cells were grown in an incubator with 5% CO₂ at 37°C.

SUNE-1 cells were divided into four groups: Control (without treatment), small interfering (si)RNA-negative control (siNC; cells transfected with 5 µg non-targeting siRNAs), si1-FOXG1 (cells transfected with 5 µg si1-FOXG1; 5′-GCCCTTCAGTTCAGGTACAAT-3′) and si2-FOXG1 (cells transfected with 5 µg si2-FOXG1; 5′-GGCTGTTTACCCACAATGAAA-3′). C666-1 cells were divided into three groups: Control (without treatment), pcDNA3.1-NC (cells transfected with 4 µg pcDNA3.1-NC vector) and pcDNA3.1-FOXG1 (cells transfected with 4 µg pcDNA3.1-FOXG1 vector). Cells (1×10⁵ cells/well) were seeded into 12-well plates and transfected using Lipofectamine^® 3000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C. The pcDNA3.1-FOXG1, siRNA-FOXG1 and NCs were purchased from Guangzhou RiboBio Biotech Co., Ltd. At 48 h post-transfection, subsequent experiments were performed.

The relative mRNA expression level of FOXG1 and mitochondrial DNA (mtDNA) copy number was measured via RT-qPCR. TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) was used to extract total RNA from tissues and cells, and a Prime-Script™ reverse transcription kit (Takara Bio, Inc.) was used to synthesize cDNA according to the manufacturer's instructions. The SYBR-Green PCR kit (Takara Bio, Inc.) was used to perform RT-qPCR on a 7500 Fast Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.). The RT-qPCR procedure was as follows: Initial denaturation at 94°C for 2 min, followed by 40 cycles of denaturation at 94°C for 30 sec, annealing at 47°C for 1 min (FOXG1) or 47°C for 30 sec (D-Loop), elongation at 72°C for 30 sec and final extension at 72°C for 5 min. The relative mRNA expression level of FOXG1 was measured using the 2^−ΔΔCq method (21). GAPDH was used as internal controls for FOXG1. The mtDNA copy number was calculated as the ratio of mitochondrial D-Loop to 18S rRNA (22). The primer sequences (Invitrogen; Thermo Fisher Scientific, Inc.) for RT-qPCR are shown in Table I.

RIPA buffer (Cell Signaling Technology, Inc.) was used to obtain total protein from tissues or cells. A BCA assay kit (Pierce; Thermo Fisher Scientific, Inc.) was used to determine the protein concentration. Protein samples (30 µg) were separated via 10% SDS-PAGE and then transferred onto PVDF membranes. The PVDF membranes were blocked with 5% non-fat milk in 0.1% TBS-Tween-20 for 1 h at room temperature. Next, the PVDF membranes were incubated with primary antibodies [FOXG1, cat. no. ab196868; N-cadherin, cat. no. ab18203; Snail, cat. no. ab229701; caspase-9, cat. no. ab219590; heat shock protein (HSP) 60, cat. no. ab190828; 1:1,000, all from Abcam; Bax, cat. no. 5023; Bcl-2, cat. no. 3498; cleaved poly(ADP-ribose) polymerase 1 (PARP), cat. no. 9185; E-cadherin, cat. no. 3195; caspase-3, cat. no. 14220; caspase-8, cat. no. 4790; succinate dehydrogenase complex flavoprotein subunit A (SDHA), cat. no. 11998; pyruvate dehydrogenase (PDH), cat. no. 3205; cleaved caspase-3, cat. no. 9654; cleaved caspase-8, cat. no. 9496; cleaved caspase-9, cat. no. 20750; 1:1,000, all from Cell Signaling Technology, Inc.; PARP, cat. no. SAB4500487; 1:1,000, Sigma-Aldrich; Merck KGaA) overnight at 4°C. The PVDF membranes were then incubated with horseradish peroxidase-conjugated goat polyclonal anti-rabbit IgG secondary antibodies (cat. no. ab150077; 1:2,000; Abcam) for 1 h at room temperature. Finally, the protein bands were visualized using an ECL reagent (Pierce; Thermo Fisher Scientific, Inc.). The intensity of protein bands were quantified using the Image Lab Software (V3.0; Bio-Rad Laboratories, Inc.).

SUNE-1 and C666-1 cells (1×10⁵ cells/well) were seeded in 96-well plates. Next, the cells were incubated in fresh RPMI-1640 medium for 24, 48, 72 or 96 h at 37°C. Then, MTT solution (10 µl) was added to each well, and the cells were cultured for 4 h in the incubator at 37°C. Subsequently, the cells were incubated with DMSO (150 µl) for 15 min at 37°C. A microplate reader (Bio-Rad Laboratories, Inc.) was used to measure the optical density value at 450 nm.

An EdU labelling/detection kit (Guangzhou RiboBio Co., Ltd.) was used to assess cell proliferation. Briefly, after transfection for 48 h, the cells were incubated with 50 µM EdU labelling medium for 2 h at 37°C. Next, the cells were fixed with 4% paraformaldehyde for 25 min at room temperature and treated with 0.5% Triton X-100 for 15 min at room temperature. Then, the cells were incubated with 1X Apollo reaction reagents (15 µl) for 25 min at room temperature and stained with DAPI solution for 10 min at room temperature. Finally, the number of EdU-positive cells was examined under a fluorescent microscope at a magnification of ×100, and the percentage of EdU-positive cells was counted from five random fields.

After transfection for 48 h, the cells (1×106 cells/well) were seeded in 6-well plates and cultured in RPMI-1640 medium containing 10% FBS to reach 90% confluence. A sterile 200-µl pipette tip was used to create wounded monolayers, and the cells were cultured in serum-free medium for 48 h at 37°C. Images of the monolayer wound were imaged at 0 and 48 h under an Olympus CKX53 microscope (Olympus Corporation) at a magnification of ×200, and the migratory ability was analyzed using ImageJ software (V1.8.0.112; National Institutes of Health) from three randomly chosen fields.

The cell invasive ability was detected using 24-well Transwell chambers (pre-coated with Matrigel for 1 h at 37°C; Corning, Inc.). In brief, after transfection for 48 h, cells (2×10⁴ cells/well) were resuspended in 200 µl serum-free medium and then added into the upper chambers. Next, 500 µl RPMI-1640 medium containing 20% FBS was added into the lower chambers. After incubation for 48 h at 37°C, cells on the upper surface were wiped using a cotton swab, and the cells were fixed with 4% paraformaldehyde for 10 min at room temperature and stained with 0.1% crystal violet for 15 min at room temperature. Finally, the number of invasive cells was counted from three random fields under an inverted light microscope at a magnification of ×200.

After transfection for 48 h, the Annexin V-FITC Apoptosis Detection kit (Sigma-Aldrich; Merck KGaA) was used to detect cell apoptosis. Cells were harvested with 0.25% trypsin and resuspended in binding buffer. Subsequently, the cells were incubated with PI (5 µl) and Annexin V-FITC (2.5 µl) for 20 min at room temperature in the dark. The cells were washed with cold PBS, and then flow cytometry (FACScan™; BD Biosciences) equipped with CellQuest software (V6.0; BD Biosciences) was conducted to calculate the proportion of apoptotic cells.

The ATP/ADP ratio was measured using the ADP/ATP ratio assay kit (Abcam). After transfection for 48 h, the cells (5×10⁴) were incubated with nucleotide-releasing buffer (200 µl) for 10 min at room temperature. Next, the cells were treated with ATP monitoring enzyme (10 µl) at room temperature for 10 min and ATP levels were measured immediately (A). ADP levels were measured (B), then cells were treated with ADP converting enzyme (10 µl) at room temperature for 5 min and ADP levels were measured (C). Finally, the ATP/ADP ratio was calculated as follows: A/(C-B).

After transfection for 48 h, the cells (5×10⁵) were grown in a medium containing 10 µg/ml 5,5′,6,6 -Tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolylcarbocyanine iodide (Thermo Fisher Scientific, Inc.). After incubation for 30 min at 37°C in the dark, the cells were washed using PBS and flow cytometry (FACScan™; BD Biosciences) equipped with CellQuest software (V6.0; BD Biosciences) was used to detect changes in MMP.

All experiments were repeated ≥3 times, and the data are expressed as the mean ± SD. GraphPad Prism 7.0 (GraphPad Software, Inc.) was used to perform statistical analyses. χ² test was used to analyze the association between FOXG1 expression and clinic characteristics in patients with NPC. An unpaired Student's t-test was used for comparison between two groups, and one-way ANOVA with Tukey's post-hoc test was used for comparison among ≥3 groups. P<0.05 was considered to indicate a statistically significant difference.

Analysis of GEO dataset (GSE12452) revealed that FOXG1 expression in NPC tissues was upregulated compared with that in normal tissues (Fig. 1A). To examine the function of FOXG1 in NPC progression, FOXG1 expression was first measured in NPC tissues and para-carcinoma tissues (Normal) via RT-qPCR and IHC. As shown in Fig. 1B and C, FOXG1 expression in NPC tissues was higher compared with that in normal tissues. According to the median mRNA expression of FOXG1, patients with NPC were divided into low and high expression groups. It was found that FOXG1 expression was associated with distant metastasis and TNM stage, while there was no significant association with sex, smoking, EBV infection and age (Table II).

Next, FOXG1 expression in NPC cells (C666-1 and SUNE-1) and NP69 cells was evaluated using RT-qPCR and western blotting. As presented in Fig. 1D and E, FOXG1 expression was upregulated in NPC cells compared with that in NP69 cells.

To investigate the effects of FOXG1 in NPC, SUNE-1 and C666-1 cells were transfected with siRNA-FOXG1 and pcDNA3.1-FOXG1, respectively. Transfection efficiency was assessed using RT-qPCR. As presented in Fig. 2A, FOXG1 expression in the si1-FOXG1 and si2-FOXG1groups was lower compared with that in the control and si-NC groups, and FOXG1 expression in the pcDNA3.1-FOXG1 group was higher compared with that in the control and pcDNA3.1-NC groups. These results indicated that transfection had been successful. Subsequently, cell proliferation was detected using MTT and EdU assays. It was found that knockdown of FOXG1 inhibited the proliferation of SUNE-1 cells compared with the control and si-NC groups, while overexpression of FOXG1 promoted the proliferation of C666-1 cells compared with the control and pcDNA3.1-NC groups (Fig. 2B and C).

To investigate the role of FOXG1 in migration and invasion of NPC cells, cells were detected via wound healing and Transwell assays. The results demonstrated that knockdown of FOXG1 inhibited the migration and invasion of SUNE-1 cells, whereas the overexpression of FOXG1 promoted the migration and invasion of C666-1 cells (Fig. 3A and B).

Subsequently, western blotting was conducted to detect the expression levels of EMT-related proteins (N-cadherin, Snail and E-cadherin) to further determine the molecular mechanism mediating the aggressive effect of FOXG1. It was identified that knockdown of FOXG1 decreased the protein expression levels of N-cadherin and Snail in SUNE-1 cells and increased the protein expression level of E-cadherin (Fig. 3C). Conversely, overexpression of FOXG1 increased the protein expression levels of N-cadherin and Snail in C666-1 cells and decreased E-cadherin protein expression.

To observe the effect of FOXG1 on cell apoptosis, flow cytometry analysis was performed on C666-1 and SUNE-1 cells. The results (Fig. 4A) demonstrated that knockdown of FOXG1 promoted the apoptosis of SUNE-1 cells, whereas overexpression of FOXG1 inhibited the apoptosis of C666-1 cells. Next, western blotting was performed to detect the expression levels of apoptosis-related proteins (Bax, Bcl-2, PARP, cleaved PARP, cleaved caspase-3, cleaved caspase-8, cleaved caspase-9, caspase-3, caspase-8 and caspase-9) in C666-1 and SUNE-1 cells. As shown in Fig. 4B, knockdown of FOXG1 increased the expression levels of Bax/Bcl-2, PARP, cleaved PARP, cleaved caspase-3, cleaved caspase-8, cleaved caspase-9, caspase-3, caspase-8 and caspase-9 in SUNE-1 cells. Notably, the opposite results were observed in the C666-1 cells with FOXG1 overexpression.

Mitochondrial dysfunction is reported to be associated with cancer cell death (19). To investigate whether FOXG1 was involved in the regulation of NPC mitochondrial function, the copy number of mtDNA was detected via RT-qPCR. As shown in Fig. 5A, knockdown of FOXG1 reduced the mtDNA copy number, whereas overexpression of FOXG1 increased the mtDNA copy number. Next, the ATP/ADP ratio was measured, and it was found that knockdown of FOXG1 resulted in a decreased ATP/ADP ratio, but overexpression of FOXG1 resulted in an increased ATP/ADP ratio (Fig. 5B). Moreover, knockdown of FOXG1 decreased the Red/Green ratio (indicative of MMP), while overexpression of FOXG1 increased the Red/Green ratio (Fig. 5C). Notably, the western blotting results revealed that knockdown of FOXG1 decreased the expression levels of mitochondrial markers (SDHA, HSP60 and PDH) in SUNE-1 cells, while the opposite results were observed in C666-1 cells with FOXG1 overexpression (Fig. 5D).